We plan to determine the incidence and type of dementia in persons in the 9th through the 11th decades of life, the oldest-old. In this most rapidly growing portion of the population, little is known about the epidemiology and clinical manifestations of dementia. The Framingham Study cohort, now includes 667 men and women, ages 78 to 104 years (mean age 85.1) documented to be free of dementia. The study of dementia in Framingham has been underway since 1976 when a comprehensive neuropsychological battery was administered and a dementia-free cohort was identified. We have since identified new cases of dementia by means of Mini-Mental State Exam (MMSE) screening, administered biennially, followed by a comprehensive neurological and neuropsychological evaluation. We will re-administer the neuropsychology battery to most of the 667 dementia-free survivors and estimate approximately 182 subjects will consent to an MRI scan of the brain. Measures of cognitive performance and decline, since the neuropsychology battery in 1976-78, will be related to: semiquantitative measures of total and regional brain volumes, white matter hyperintensity (WMHI) volume, silent cerebral infarcts (SCI) and stroke volume on MR scans. Utilizing the existing brain donation program, systematic semiquantitative neuropathological study of the brains of these subjects will be related to MRI, cognitive performance and decline, and, previously measured risk factors. Using data collected prospectively and systematically over the course of the Framingham Study, as recently as 2 years before and as remotely as 50 years ago, we plan to relate lifestyle variables, medical information, laboratory measures, documented medical illness and cognitive functioning to the incidence of probable AD and vascular dementia. We will identify all new cases of dementia in these elderly survivors of this general population sample during the next five years. Dementia may result more often from vascular disease than from AD in persons of advanced age. Based on MRI, neurological and neuropsychological evaluations, (and in a small number based on neuropathologic study) we will be able to confirm the diagnosis and determine the mechanism underlying the dementia in these subjects. We will also be in a unique position to relate precursors of, and prior cardiovascular disease to dementia including vascular dementia and to determine the contribution of vascular factors to the dementia of AD.